Dyeing recipes prediction optimization for yarn color matching based on conjugate gradient method

In order to improve the accuracy and stability of the predicted yarn dyeing recipes,the conjugate gradient-based optimization method of dyeing recipes prediction is proposed for the color matching on the basis of the Kubelka-Munk theory,CIE 1976 L* a* b* uniform color space and the CIELAB color difference formula.The color matching objective function with constraints for dyeing recipes prediction optimization is deduced.Finally,through the self-developed color measuring matching system and Minolta CM2600D(a spectral color measurement instrument) with D65 light source,a lot of experiments are carried out for yarn color measuring and matching.The experimental results show that the colors obtained by calculated dyeing recipes using the above-mentioned method are close to the standard ones of yarn samples and the color difference between the optimization results and actual dying recipes is little.The optimization results have nothing to do with the initial value,and the calculation process is stable and reliable,and has strong convergence and high reproducibility.